From: Nilay Shroff <nilay@linux.ibm.com>
To: Yu Kuai <yukuai1@huaweicloud.com>, ming.lei@redhat.com, axboe@kernel.dk
Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
yukuai3@huawei.com, yi.zhang@huawei.com, yangerkun@huawei.com,
johnny.chenyi@huawei.com
Subject: Re: [PATCH for-6.18/block 04/10] blk-mq: convert to serialize updating nr_requests with update_nr_hwq_lock
Date: Tue, 9 Sep 2025 11:59:03 +0530 [thread overview]
Message-ID: <9708abeb-7677-4c0e-931b-7ca5fe0a0242@linux.ibm.com> (raw)
In-Reply-To: <20250908061533.3062917-5-yukuai1@huaweicloud.com>
On 9/8/25 11:45 AM, Yu Kuai wrote:
> From: Yu Kuai <yukuai3@huawei.com>
>
> request_queue->nr_requests can be changed by:
>
> a) switching elevator by update nr_hw_queues
> b) switching elevator by elevator sysfs attribute
> c) configue queue sysfs attribute nr_requests
>
> Current lock order is:
>
> 1) update_nr_hwq_lock, case a,b
> 2) freeze_queue
> 3) elevator_lock, cas a,b,c
>
> And update nr_requests is seriablized by elevator_lock() already,
> however, in the case c), we'll have to allocate new sched_tags if
> nr_requests grow, and do this with elevator_lock held and queue
> freezed has the risk of deadlock.
>
> Hence use update_nr_hwq_lock instead, make it possible to allocate
> memory if tags grow, meanwhile also prevent nr_requests to be changed
> concurrently.
>
> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
> ---
> block/blk-sysfs.c | 12 +++++++++---
> 1 file changed, 9 insertions(+), 3 deletions(-)
>
> diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
> index f99519f7a820..7ea15bf68b4b 100644
> --- a/block/blk-sysfs.c
> +++ b/block/blk-sysfs.c
> @@ -68,13 +68,14 @@ queue_requests_store(struct gendisk *disk, const char *page, size_t count)
> int ret, err;
> unsigned int memflags;
> struct request_queue *q = disk->queue;
> + struct blk_mq_tag_set *set = q->tag_set;
>
> ret = queue_var_store(&nr, page, count);
> if (ret < 0)
> return ret;
>
> - memflags = blk_mq_freeze_queue(q);
> - mutex_lock(&q->elevator_lock);
> + /* serialize updating nr_requests with switching elevator */
> + down_write(&set->update_nr_hwq_lock);
>
For serializing update of nr_requests with switching elevator,
we should use disable_elv_switch(). So with this change we
don't need to acquire ->update_nr_hwq_lock in writer context
while running blk_mq_update_nr_requests but instead it can run
acquiring ->update_nr_hwq_lock in the reader context.
So the code flow should be,
disable_elv_switch => this would set QUEUE_FLAG_NO_ELV_SWITCH
...
down_read ->update_nr_hwq_lock
acquire ->freeze_lock
acquire ->elevator_lock;
...
...
release ->elevator_lock;
release ->freeze_lock
clear QUEUE_FLAG_NO_ELV_SWITCH
up_read ->update_nr_hwq_lock
Thanks,
--Nilay
next prev parent reply other threads:[~2025-09-09 6:29 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-08 6:15 [PATCH for-6.18/block 00/10] cleanup and fixes for updating nr_requests Yu Kuai
2025-09-08 6:15 ` [PATCH for-6.18/block 01/10] blk-mq: remove useless checking in queue_requests_store() Yu Kuai
2025-09-09 11:34 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 02/10] blk-mq: remove useless checkings in blk_mq_update_nr_requests() Yu Kuai
2025-09-09 11:35 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 03/10] blk-mq: check invalid nr_requests in queue_requests_store() Yu Kuai
2025-09-09 11:36 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 04/10] blk-mq: convert to serialize updating nr_requests with update_nr_hwq_lock Yu Kuai
2025-09-09 6:29 ` Nilay Shroff [this message]
2025-09-09 6:38 ` Yu Kuai
2025-09-09 6:52 ` Nilay Shroff
2025-09-09 7:16 ` Yu Kuai
2025-09-09 9:26 ` Nilay Shroff
2025-09-09 9:36 ` Yu Kuai
2025-09-09 10:11 ` Nilay Shroff
2025-09-09 10:42 ` Yu Kuai
2025-09-09 11:32 ` Nilay Shroff
2025-09-09 11:40 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 05/10] blk-mq: cleanup shared tags case in blk_mq_update_nr_requests() Yu Kuai
2025-09-09 11:58 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 06/10] blk-mq: split bitmap grow and resize " Yu Kuai
2025-09-09 12:18 ` Nilay Shroff
2025-09-09 16:39 ` Yu Kuai
2025-09-10 6:30 ` Nilay Shroff
2025-09-10 6:42 ` Yu Kuai
2025-09-08 6:15 ` [PATCH for-6.18/block 07/10] blk-mq-sched: add new parameter nr_requests in blk_mq_alloc_sched_tags() Yu Kuai
2025-09-09 12:19 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 08/10] blk-mq: fix potential deadlock while nr_requests grown Yu Kuai
2025-09-09 6:39 ` Nilay Shroff
2025-09-09 7:37 ` Yu Kuai
2025-09-09 9:36 ` Nilay Shroff
2025-09-09 12:21 ` Nilay Shroff
2025-09-10 7:46 ` Yu Kuai
2025-09-08 6:15 ` [PATCH for-6.18/block 09/10] blk-mq: remove blk_mq_tag_update_depth() Yu Kuai
2025-09-09 12:35 ` Nilay Shroff
2025-09-08 6:15 ` [PATCH for-6.18/block 10/10] blk-mq: fix stale nr_requests documentation Yu Kuai
2025-09-09 12:35 ` Nilay Shroff
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9708abeb-7677-4c0e-931b-7ca5fe0a0242@linux.ibm.com \
--to=nilay@linux.ibm.com \
--cc=axboe@kernel.dk \
--cc=johnny.chenyi@huawei.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ming.lei@redhat.com \
--cc=yangerkun@huawei.com \
--cc=yi.zhang@huawei.com \
--cc=yukuai1@huaweicloud.com \
--cc=yukuai3@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox